555863

research-article2014
JIVXXX10.1177/0886260514555863Journal of Interpersonal ViolenceEarnest and Brady

Article
Journal of Interpersonal Violence
2016, Vol. 31(3) 383­–406
Dating Violence © The Author(s) 2014
Reprints and permissions:
Victimization Among sagepub.com/journalsPermissions.nav
DOI: 10.1177/0886260514555863
High School Students in jiv.sagepub.com

Minnesota: Associations
With Family Violence,
Unsafe Schools, and
Resources for Support

Alicia A. Earnest, MPH1

and Sonya S. Brady, PhD1

Abstract
The present study examines whether being a victim of violence by an adult in
the household, witnessing intra-familial physical violence, and feeling unsafe
at school are associated with physical dating violence victimization. It also
examines whether extracurricular activity involvement and perceived care
by parents, teachers, and friends attenuate those relationships, consistent
with a stress-buffering model. Participants were 75,590 ninth-and twelfth-
grade students (51% female, 77% White, 24% receiving free/reduced price
lunch) who completed the 2010 Minnesota Student Survey. Overall, 8.5%
of students reported being victims of dating violence. Significant differences
were found by gender, grade, ethnicity, and free/reduced price lunch status.
Logistic regression analyses demonstrated that being a victim of violence
by an adult in the household, witnessing intra-familial physical violence,
feeling unsafe at school, and low perceived care by parents were strongly
associated with dating violence victimization. Associations of moderate

1University of Minnesota School of Public Health, Minneapolis, MN, USA

Corresponding Author:
Sonya S. Brady, Division of Epidemiology & Community Health, University of Minnesota
School of Public Health, 1300 S. 2nd Street, Suite 300, Minneapolis, MN 55454, USA.
Email: ssbrady@umn.edu
384 Journal of Interpersonal Violence 31(3)

strength were found for low perceived care by teachers and friends. Little
to no extracurricular activity involvement was weakly associated with
dating violence victimization. Attenuating effects of perceived care and
extracurricular activity involvement on associations between risk factors
(victimization by a family adult, witnessing intra-familial violence, feeling
unsafe at school) and dating violence victimization were smaller in magnitude
than main effects. Findings are thus more consistent with an additive model
of risk and protective factors in relation to dating violence victimization
than a stress-buffering model. Health promotion efforts should attempt to
minimize family violence exposure, create safer school environments, and
encourage parental involvement and support.

Keywords
dating violence, adolescents, victimization, social support, school safety

Nearly 10% of adolescents in the United States are victims of physical dating
violence (Centers for Disease Control and Prevention, 2012). Physical dating
violence victimization encompasses non-sexual bodily harm from a romantic
partner such as hitting, pushing, kicking, or slapping (Teten, Ball, Valle,
Noonan, & Rosenbluth, 2009). An ecological systems perspective suggests that
an adolescent’s risk for dating violence victimization is a function of risk and
protective factors that surround the adolescent and that may be proximal or
distal; factors include family structure and processes; peer and partner charac-
teristics; neighborhood, school, and community characteristics; societal norms
conveyed through media; and broader cultural norms and practices (Connolly,
Friedlander, Pepler, Craig, & Laporte, 2010; Jain, Buka, Subramanian, &
Molnar, 2010). While all sources of influence are important, experiences in the
home and school may be particularly likely to shape young people’s attitudes
about the acceptability of violence (Chiodo, Wolfe, Crooks, Hughes, & Jaffe,
2009; Foshee, Benefield, Ennett, Bauman, & Suchindran, 2004; Maas, Fleming,
Herrenkohl, & Catalano, 2010; Malik, Sorenson, & Aneshensel, 1997). The
present study examines risk and protective factors in the adolescent home and
school environment in relationship to dating violence victimization.
Experiences of violence in the home may predispose subsets of adoles-
cents to dating violence victimization by modeling violence in relationships
as normative behavior (Bandura, 1971). Bandura’s social learning theory has
been the basis for subsequent theoretical work on intergenerational transmis-
sion of violence (e.g., Egeland, 1993; Kalmuss, 1984). Prospective studies
(Foshee et al., 2004; Maas et al., 2010; Tyler, Brownridge, & Melander, 2011)
Earnest and Brady 385

and cross-sectional studies (Laporte, Jiang, Pepler, & Chamberland, 2011;

Malik et al., 1997; E. Miller et al., 2011) have shown that adolescents who
experience physical abuse by a parent or caregiver are more likely to be vic-
timized by dating violence than those who do not experience this abuse.
Research on dating violence victimization among adolescents who witness
physical abuse between adults in the home is mixed. At least three studies
have linked witnessing physical abuse between adults in the home to adoles-
cent dating violence victimization (Maas et al., 2010; Malik et al., 1997; E.
Miller et al., 2011); in contrast, at least two studies have found no relationship
(Foshee et al., 2004; Gagne, Lavoie, & Hebert, 2005).
Adolescents may also experience violence and threats to their safety at
school. Studies have not examined whether perceived school safety is linked
to dating violence victimization. Some research has examined links between
perceived school safety and harassment, another form of relationship vio-
lence. Students who report feeling unsafe at school are more likely to report
being sexually harassed by another teen or classmate(s) (Chiodo et al., 2009)
and being bullied by other students at school (Beran & Tutty, 2002; Glew,
Fan, Katon, & Rivara, 2008). These studies were cross-sectional; while it is
likely that being bullied or harassed makes students feel unsafe at school, it is
also possible that unsafe school environments normalize and facilitate further
bullying or harassment, leading to greater incidence of victimization among
students.
Research has also demonstrated links between being victimized by bully-
ing or sexual harassment and being victimized by dating violence victimiza-
tion (Chiodo et al., 2009; Espelage & Holt, 2007; Gagne et al., 2005). Chiodo
et al. (2009) found that students who were victims of sexual harassment in
Grade 9 were more likely to report physical dating violence victimization 2
years later. In their cross-sectional study, Gagne et al. (2005) explicitly
excluded dating partners from assessments of sexual harassment and bullying
by peers when assessing links to dating violence victimization. This is impor-
tant because it strengthens the interpretation that one type of violence inflicted
by peers (i.e., sexual harassment or bullying) may make adolescents more
vulnerable to dating violence victimization. These studies also support the
idea that unsafe school environments can normalize violence and lead to
greater likelihood of victimization by dating violence.

Resources for Resilience in the Context of Risk

Not all adolescents who are exposed to violence in the home or school
become involved in violent dating relationships. The stress-buffering hypoth-
esis suggests that social support can minimize the negative impact of stressful
386 Journal of Interpersonal Violence 31(3)

life experiences on health outcomes by providing psychological and material

resources to cope with stress (Cohen & Wills, 1985). Applying this hypothe-
sis to dating violence, adolescents who experience violence in the home or
school may be less likely to form violent relationships or to tolerate violence
in dating relationships if they have resources for social support.
At least two studies have examined whether social support and family
violence exposure interact to influence dating violence victimization. In a
cross-sectional survey of college students, L. G. Simons, Simons, Lei,
Hancock, and Fincham (2012) examined the potential interaction between
maternal and/or paternal hostility, defined by experiences of parental verbal
or emotional abuse, and maternal and/or paternal warmth in relation to dating
violence victimization. Findings demonstrated that maternal warmth inter-
acted with maternal hostility such that the association between maternal hos-
tility and dating violence victimization was stronger in the presence of
maternal warmth. The same was true for the interaction between paternal
warmth and paternal hostility. These results were significant for female but
not male students. L. G. Simons et al. (2012) also found that interparental
warmth amplified the effect of witnessing interparental hostility (defined as
verbal and/or physical abuse between parents) on male and female adolescent
dating violence victimization. Findings of this study contradict the stress-
buffering hypothesis; instead of attenuating the relationship between family
hostility and dating violence victimization, perceptions of warmth actually
exacerbated the relationship. L. G. Simons et al. (2012) posited that parental
demonstration of both hostility and warmth within their romantic relationship
led the child to conclude that aggression is a normal part of romantic relation-
ships. L. G. Simons et al. (2012) did not examine whether maternal and/or
paternal warmth directed toward the adolescent moderated the effect of inter-
parental hostility on dating violence victimization.
In the second study, Levendosky, Huth-Bocks, and Semel (2002) found
that greater social support from friends was associated with less abuse from
dating partners among adolescents whose mothers reported low amounts of
domestic violence (defined as being physically harmed or threatened by her
intimate partner). In contrast, greater social support from friends was associ-
ated with greater abuse from dating partners among adolescents whose moth-
ers reported high amounts of domestic violence. The authors posited that
adolescents exposed to high amounts of domestic violence were surrounded
by more violent peer networks in which violent relationships were seen as the
norm; social support may have made adolescents more receptive to this norm.
Levendosky et al. did not examine whether domestic violence exposure was
associated with dating partner abuse within groups of adolescents reporting
low or high social support from friends. This approach is of interest because
Earnest and Brady 387

it would test whether a resource for resilience (social support from friends)
appeared to attenuate the impact of a risk factor (domestic violence exposure)
on dating violence victimization. In addition to the L. G. Simons et al. (2012)
and Levendosky et al. (2002) studies, Tyler et al. (2011) found independent
effects of physical abuse by a parent or caregiver and low adolescent-per-
ceived parental warmth on dating violence victimization in a longitudinal
study of middle and high school adolescents. However, Tyler and colleagues
(2011) did not test for an interaction between physical abuse and parental
warmth in relation to dating violence victimization.
Tajima, Herrenkohl, Moylan, and Derr (2011) found support for the stress-
buffering hypothesis with respect to negative adolescent outcomes other than
dating violence victimization. In a longitudinal sample of adolescents, posi-
tive youth perceptions of parents’ responsiveness and attentiveness to emo-
tional needs attenuated the effect of witnessing interparental physical violence
on both running away from home and teenage pregnancy. In addition, peer
communication and peer trust attenuated the effect of witnessing interparen-
tal physical violence on running away from home, dropping out of high
school, and depression.
In addition to direct support from others, participation in extracurricular
activities may conceivably act as a buffer for adolescents who are exposed to
violence at home or school. This conceptualization is consistent with the idea
that opportunities for social support are enhanced through increasing the size
of one’s social network (Cohen & Wills, 1985). No studies have examined
whether extracurricular activity involvement is directly associated with dat-
ing violence victimization, nor whether it moderates associations between
violence in the home or perceptions of school safety and dating violence
victimization. Research has demonstrated that adolescents’ involvement in
extracurricular activities can reduce the likelihood of engaging in a variety of
risk behaviors, including tobacco and other drug use, skipping school, drop-
ping out of school, and sexual risk-taking (Cooley, Henriksen, Van Nelson, &
Thompson, 1995; Eccles & Barber, 1999; Eccles, Barber, Stone, & Hunt,
2003; Mahoney & Cairns, 1997; Mahoney & Stattin, 2000; McNeal, 1995; K.
E. Miller, Sabo, Farrell, Barnes, & Melnick, 1999; Youniss, Yates, & Su,
1997). In addition, adolescents’ involvement in extracurricular activities has
been associated with positive outcomes such as high self-esteem, emotion
regulation, mental health, and healthy relationships (Barber, Eccles, & Stone,
2001; Fredricks & Eccles, 2006; Larson, Hansen, & Moneta, 2006; Marsh &
Kleitman, 2003). The number of different activities in which an adolescent
participates has been associated with positive academic and behavioral out-
comes such as higher grades and years of schooling, civic engagement, and
lower marijuana use (Fredricks & Eccles, 2006). Thus, research has
388 Journal of Interpersonal Violence 31(3)

demonstrated that extracurricular activity involvement is directly associated

with lower behavioral risk and positive characteristics that may serve as
resources in the context of risk. It remains to be seen whether extracurricular
activity involvement attenuates the relationship between violence in the
home or school and dating violence victimization. It is also important to note
that extracurricular activities could conceivably enhance a relationship
between violence in the home or school and dating violence victimization.
Characteristics of peer networks that are accessed through extracurricular
activity involvement can vary. Social support from a new network may be
“packaged” with negative peer characteristics, such as tolerance or accep-
tance of violence; this could potentially enhance risk for dating violence vic-
timization, consistent with the findings of Levendosky et al. (2002).
In the absence of an interaction between stressors and social support, it is
possible that each factor may exert independent risk and protective effects on
dating violence victimization, respectively. Such an additive model of risk
and protection would be consistent with the manner in which most social
ecological determinants of health among youth are examined in the literature
(e.g., Masten, Cutuli, Herbers, & Reed, 2009; Masten & Powell, 2003;
Pollard, Hawkins, & Arthur, 1999).

Demographic Factors Linked With Dating Violence

Victimization
Along with family violence, unsafe schools, and resources for resilience, it is
important to consider several demographic characteristics that increase risk
for dating violence victimization, including older age (Malik et al., 1997;
Spriggs, Halpern, Herring, & Schoenbach, 2009), African American, Latino,
or Native American heritage (Halpern, Spriggs, Martin, & Kupper, 2009;
Malik et al., 1997; Spriggs et al., 2009), socioeconomic disadvantage
(Aldarondo & Sugarman, 1996; Breiding, Black, & Ryan, 2008; Cunradi,
Caetano, & Schafer, 2002; Foshee et al., 2004; O’Keefe & Treister, 1998;
Spriggs et al., 2009; Tjaden & Thoennes, 2000), and gender (Field & Caetano,
2005; Foshee et al., 2004; Halpern et al., 2009; Maas et al., 2010; Malik et al.,
1997; O’Keefe & Treister, 1998; O’Leary, Smith Slep, Avery-Leaf, &
Cascardi, 2008; Tjaden & Thoennes, 2000; Windle & Mrug, 2009). Some
studies report that the prevalence of dating violence victimization is higher
among male than female adolescents (Malik et al., 1997; O’Keefe & Treister,
1998), while others report the opposite (Halpern et al., 2009; Maas et al.,
2010). Females tend to suffer more severe violence than males (Field &
Caetano, 2005; Foshee et al., 2004; O’Keefe & Treister, 1998; Tjaden &
Earnest and Brady 389

Thoennes, 2000), but more often perpetrate milder forms of violence against
male partners (O’Leary et al., 2008; Windle & Mrug, 2009).

Hypotheses
The present study examines how risk and protective factors in the home and
school may be linked to dating violence victimization. Each risk factor (wit-
nessing intra-familial physical violence, being a victim of physical violence
by an adult in the family, feeling unsafe at school) was hypothesized to
increase risk for dating violence victimization independent of gender, grade,
race, and free or reduced price lunch status. It was hypothesized that social
support from different sources (parents, friends, and teachers) and high levels
of extracurricular activity involvement would have direct protective effects
against dating violence victimization. Finally, it was hypothesized that asso-
ciations between risk factors and dating violence victimization would be
attenuated by social support and by level of participation in extracurricular
activities, consistent with a stress-buffering model.

Method
Procedure
This research utilizes cross-sectional, de-identified data from the 2010
Minnesota Student Survey and was therefore deemed exempt from review by
the authors’ Institutional Review Board. Data were collected through a col-
laboration between Minnesota schools and several state departments.
Respondents in 2010 included 6th-, 9th-, and 12th-grade students in public
schools, including charter and tribal schools. Questions on the survey assessed
student demographics and health-related experiences, perceptions, and
behaviors. The survey was administered in English during one class period
and consisted of a self-administered paper-and-pencil scan sheet with a sepa-
rate packet of questions. Student participation was voluntary; all participat-
ing students provided assent. The parental consent procedure varied by
school district, with most using opt-out, passive consent. All surveys were
anonymous. While all public school districts in Minnesota were invited to
participate, administrative leadership from 295 out of 335 school districts
(88%) elected to participate. The present study only includes 9th- and 12th-
grade students, who were asked questions about dating violence exposure.
Participation rates varied by grade (9th grade, 75%; 12th grade, 59%). It is
not known whether students who did and did not participate differed on any
of the variables examined.
390 Journal of Interpersonal Violence 31(3)

Participants
Students who did not answer all questions of interest for the study (10%)
were excluded, yielding a final N of 75,590. Table 1 presents the composition
of the study sample with respect to demographics.

Measures
Outcome. Dating violence victimization was assessed by the yes/no question,
“Has someone you were going out with ever hit you, hurt you, threatened you
or made you feel afraid?”

Risk factors. Feeling unsafe at school was assessed by a single question: “I

feel safe at school.” Responses options were “strongly agree,” “agree,” “dis-
agree,” and “strongly disagree.” Responses of strongly agree and agree were
combined to indicate a perceived safe school environment. Responses of
strongly disagree and disagree were combined to indicate a perceived unsafe
school environment. Physical violence victimization by an adult in the home
was assessed by the yes/no question, “Has any adult in your household ever
hit you so hard or so often that you had marks or were afraid of that person?”
Witness to intra-familial physical violence was assessed by the yes/no ques-
tion, “Has anyone else in your family ever hit anyone else in the family so
hard or so often that they had marks or were afraid of that person?”

Protective factors. Students were asked “How much do you feel . . . (a) friends
care about you? (b) teachers/other adults at school care about you? (c) your
parents care about you?” Response options were “very much,” “quite a bit,”
“some,” “a little,” and “not at all.” Each source of perceived care was assessed
separately. Students who felt that the potential source of support cared about
them “very much” or “quite a bit” were considered to have high perceived
care; those who felt the potential source of support cared about them “some,”
“a little,” or “not at all” were considered to have low perceived care.
Students were asked how often they participated in eight different extra-
curricular activities: fine arts, club/community sports teams, school sports
teams, community clubs/programs, mentoring programs, religious activities,
service learning programs, and tutoring/other academic programs. Responses
were coded as follows: “not available in my community” and “never” were
coded as 0; “less than monthly” was coded as 0.01; “monthly” was coded as
0.03; “1-2 times per week” was coded as 1.5; “3-4 times per week” was
coded as 3.5; and a response of “every day” was coded as 5. Students’ answers
were then summed across all eight activities to determine how many activity
Earnest and Brady 391

Table 1. Distributions of Study Variables and Prevalence of Dating Violence

Victimization (N = 75,590).

Proportion Reporting
Percentage Dating Violence
Variable n of Sample Victimizationa
Victim of dating violence 6,426 8.50 —
Gender
Female 38,907 51.47 .11
Male 36,683 48.53 .06
Grade
12 33,410 44.20 .11
9 42,180 55.80 .07
Ethnicityb
White 58,251 77.06 .08
American Indian 807 1.07 .15
Black, African, or African 3,342 4.42 .10
American
Hispanic or Latino 2,773 3.67 .11
Asian American or Pacific 3,988 5.28 .08
Islander
More than one race or 4,745 6.28 .14
ethnicity
I do not know/no answer 1,684 2.23 .14
Free or reduced price lunch
Yes 18,282 24.19 .11
No 57,308 75.81 .08
Feels unsafe at school
Yes 4,622 6.10 .24
No 70,977 93.90 .08
Victim of physical violence by adult in home
Yes 6,834 9.04 .27
No 68,756 90.96 .07
Witness to intra-familial physical violence
Yes 7,991 10.57 .23
No 67,599 89.43 .07
Perceived care from parents
Low 7,231 9.57 .21
High 68,359 90.43 .07
Perceived care from teachers
Low 40,727 53.88 .11
High 34,863 46.12 .06
(continued)
392 Journal of Interpersonal Violence 31(3)

Table 1. (continued)

Proportion Reporting
Percentage Dating Violence
Variable n of Sample Victimizationa
Perceived care from friends
Low 17,450 23.09 .13
High 58,140 76.91 .07
Involvement in extracurricular activities
Very high 19,228 25.44 .07
High 23,439 31.01 .07
Low-moderate 17,221 22.78 .09
Little to none 15,702 20.77 .12
aChi-square tests were performed on counts. p < .0001 for all tests due to very large
sample size.
bIn subsequent regression models, the ethnicity variable was treated as dichotomous. “I don’t

know/No answer” was treated as missing. The White category was assigned the value of 0,
and all other groups were assigned the value of 1.

events they attended in a given week. The sample distribution was examined,
and four evenly distributed categories were created. Less than one activity
event per school week indicated little or no involvement, 1 to 5 activity events
indicated low to moderate involvement, more than 5 and up to 10 activity
events indicated high involvement, more than 10 activity events indicated
very high involvement.

Data Analyses
Preliminary analyses examined differences between the study sample and
those excluded due to missing data, distributions of all study variables within
the study sample, and differences in dating violence prevalence by demo-
graphics and other study variables. Logistic regression analyses were used to
test all study hypotheses. Gender, grade, ethnicity, and free or reduced price
lunch status were included as covariates in all models. Basic assumptions for
logistic regression were verified, including absence of multicollinearity,
independence of errors, and sufficient case to variable ratio; linearity in the
logit and check for influential outliers did not apply due to the categorical
nature of all independent variables.
Main effects were first examined adjusting only for covariates. Next, a
fully adjusted model (i.e., additive model) included all variables to examine
Earnest and Brady 393

potentially independent effects of different risk and protective factors on the

likelihood of dating violence victimization. The stress-buffering hypothesis
was tested by conducting a series of regression analyses in which a single
interaction term between a specific risk factor and protective factor was
entered into each model.

Results
Chi-square analyses showed that there were significant differences on all
demographic variables between the study sample and those excluded from
the study due to any missing variable (n = 8,531). Excluded students had a
higher proportion of males than the study sample (60% vs. 49%), were more
likely to be in the 9th grade (61% vs. 56%), were more likely to report a non-
White ethnicity (37% vs. 29%), and were more likely to receive free or
reduced price lunch (38% vs. 24%). Among the excluded students who
answered the dating violence victimization question (n = 4,494), more than
11% reported victimization compared with 8.5% of students who answered
all study questions.

Variable Distributions and Differences in Dating Violence

Prevalence
The overall prevalence of ever having been a victim of dating violence was
8.5% (see Table 1). The majority of students reported that they felt safe at
school (94% vs. 6%). Nine percent of students reported having been a victim
of physical violence by an adult in their home, and more than 10% reported
having witnessed intra-familial physical violence. In all, 90%, 46%, and 77%
reported high perceived care from parents, teachers, and friends, respectively.
By design, the distribution of extracurricular activity involvement was fairly
even across the four groups.
Table 1 also shows differences in prevalence of dating violence victimiza-
tion by study variables. Differences were the most pronounced for students
who felt unsafe at school, experienced violence in the home, or perceived low
care from parents. Students who reported feeling unsafe at school had a
higher prevalence than those who reported feeling safe (24% vs. 8%). Similar
results were found for students who had been physically abused by an adult
in the family (27% vs. 7%) and for students who had witnessed intra-familial
physical violence (23% vs. 7%). In general, prevalence of dating violence
victimization was higher among the low perceived care groups and among
those who reported less involvement in extracurricular activities. The
394 Journal of Interpersonal Violence 31(3)

difference by perceived care from parents was the most pronounced (21% vs.
7% for low and high perceived care, respectively).

Main Effects of Study Variables on Dating Violence

Victimization
When adjusted only for demographic variables, being a victim of physical
violence by an adult in the family had the strongest effect on dating violence
victimization (OR = 4.87; see Table 2), followed by feeling unsafe at school
(OR = 4.06) and witnessing intra-familial physical violence (OR = 3.84).
Low perceived care from parents also had a strong effect on dating violence
victimization (OR = 3.50). Weaker effects were found for low perceived care
from teachers (OR = 2.09) and friends (OR = 2.06). Little or no (OR = 1.70)
and low-moderate (OR = 1.19) extracurricular activity involvement pre-
sented a small risk for dating violence victimization compared with very high
involvement, while high involvement was weakly protective (OR = 0.92).
In the fully adjusted model, these main effects were still present. Feeling
unsafe at school had the strongest effect on dating violence victimization
(OR = 2.53; see Table 2), closely followed by being a victim of physical
violence by an adult in the family (OR = 2.48). The effects for witnessing
intra-familial physical violence (OR = 1.86), low perceived care from
parents (OR = 1.68), low perceived care from teachers (OR = 1.44), and
low perceived care from friends (OR = 1.23) were weaker. Little or no
extracurricular activity involvement (OR = 1.28) presented a small risk for
dating violence victimization when compared with very high involvement,
while high involvement (OR = 0.89) was weakly protective. Low-moderate
extracurricular activity involvement was not significantly different from
very high involvement with respect to likelihood of dating violence
victimization.

Stress-Buffering Effects of Perceived Care and Extracurricular

Activity Involvement
Tests of moderation by perceived care from parents, teachers, and friends on
the associations between risk variables and dating violence victimization
were performed (data not shown). Tests were statistically significant with
two exceptions: the interaction between perceived care from friends and
being a victim of physical violence by an adult in the family, and the interac-
tion between perceived care from friends and witnessing intra-familial phys-
ical violence. All significant interactions were plotted using predicted
Earnest and Brady 395

Table 2. Logistic Regression of Dating Violence Victimization on Risk and

Protective Factors.
Dating Violence Victimization

Adjusted for Base Model

Variables Only Fully Adjusted Additive Model

Odds Ratioa [95% CI] Odds Ratiob [95% CI]

Base model
Female gender 2.18 [2.06, 2.30] 2.27 [2.14, 2.40]
Grade 12 1.77 [1.68, 1.86] 1.93 [1.83, 2.04]
Ethnic minority 1.30 [1.22, 1.39] 1.07 [0.99, 1.14]
Free or reduced price 1.42 [1.34, 1.52] 1.10 [1.03, 1.17]
lunch
Model 1
Feels unsafe at school 4.06 [3.76, 4.39] 2.53 [2.32, 2.75]
Model 2
Victim of physical 4.87 [4.56, 5.19] 2.48 [2.29, 2.70]
violence by adult in
family
Model 3
Witness to intra-familial 3.84 [3.61, 4.10] 1.86 [1.72, 2.02]
physical violence
Model 4
Low perceived care 3.50 [3.28, 3.74] 1.68 [1.56, 1.81]
from parents
Model 5
Low perceived care 2.09 [1.97, 2.21] 1.44 [1.36, 1.54]
from teachers
Model 6
Low perceived care 2.06 [1.95, 2.19] 1.23 [1.15, 1.32]
from friends
Model 7
Involvement in extracurricular activities
  Very high 1.00 1.00
  High 0.92 [0.85, 0.99] 0.89 [0.82, 0.96]
  Low-moderate 1.19 [1.10, 1.29] 1.03 [0.95, 1.12]
   Little to none 1.70 [1.57, 1.83] 1.28 [1.18, 1.39]

Note. Odds ratios are presented from logistic regressions of dating violence victimization on predictors.
CI = confidence interval.
ap < .05 for all tests.
bp < .05 for all tests with the exceptions of ethnic minority status and low-moderate involvement in

extracurricular activities.

probabilities from each model. Figure 1 shows an example. Feeling unsafe

at school had a strong effect on dating violence victimization among groups
396 Journal of Interpersonal Violence 31(3)

Figure 1. Perceived care by parents attenuates the association between feeling

unsafe at school and dating violence victimization. Attenuation is modest compared
with the relatively strong main effect of feeling unsafe at school.

of adolescents who were high or low in perceived care from parents (repre-
sented by the dashed and solid lines, respectively). The magnitude of the
effect was somewhat stronger for those who reported low perceived care
from parents compared with those who reported high perceived care from
parents; however, the difference between groups was small (note that the
lines in Figure 1 are nearly parallel). In other words, the stress-buffering
effect of perceived care was small in magnitude given the strong main effect
of feeling unsafe at school. All other plots of significant interactions involv-
ing perceived care were similar to Figure 1 (not shown).
Tests of moderation by extracurricular activity involvement on the asso-
ciations between risk variables and dating violence victimization were per-
formed (data not shown). All interactions were statistically significant and
were plotted using predicted probabilities from each model. Figure 2 shows
the stress-buffering effect of extracurricular activity involvement on the
Earnest and Brady 397

Figure 2. In all groups, being a victim of physical violence by an adult in the home
is associated with greater likelihood of experiencing dating violence victimization.
Low to moderate and high extracurricular activity involvement have modest
attenuating effects.

relationship between physical violence victimization by an adult in the home

and dating violence victimization. Those who reported physical violence vic-
timization by an adult in the home had a higher probability of experiencing
dating violence victimization than those who had not reported physical vio-
lence victimization by an adult in the home. This relationship slightly dif-
fered by level of extracurricular activity involvement; among adolescents
with little to no extracurricular activity involvement or very high involve-
ment, physical violence victimization by an adult in the home had a more
pronounced association with dating violence victimization. However, the
stress-buffering effect of extracurricular activity involvement was small in
magnitude given the strong main effect of being a victim of physical violence
by an adult in the home. This was true for all tests of moderation involving
extracurricular activity involvement.
398 Journal of Interpersonal Violence 31(3)

Discussion
Consistent with theories that emphasize modeling of behavior (Bandura,
1971; Egeland, 1993; Kalmuss, 1984) and previous research (Foshee et al.,
2004; Laporte et al., 2011; Maas et al., 2010; Malik et al., 1997; K. E. Miller
et al., 2011; Tyler et al., 2011), being a victim of physical violence in the
home was strongly associated with adolescent dating violence victimization.
In addition, witnessing intra-familial physical violence was associated with
dating violence victimization; this, too, is consistent with theory and previous
research (Maas et al., 2010; Malik et al., 1997; E. Miller et al., 2011). Novel
findings from the present study demonstrate that feeling unsafe at school is
strongly associated with adolescent dating violence victimization. Inferences
of causality are difficult to make with respect to this association. Adolescents
who experience dating violence may feel unsafe at school because of the
potential presence of their abusive dating partner, who may be perpetrating
abuse while at school. Alternatively, students may feel unsafe at school due
to norms that facilitate victimization by peers, which may in turn increase the
likelihood of experiencing dating violence. Prospective research should
examine mechanisms that may link perceived school safety and dating vio-
lence victimization.
To a lesser extent than family violence exposure and perceived lack of
school safety, lack of resources for social support was associated with dating
violence victimization. In particular, students who perceived low care from
their parents were at greater risk for dating violence victimization than stu-
dents who perceived high care. Parents are an important source of care, pro-
tection, and affirmation, helping to nurture adolescents’ sense of self-worth
and control over their lives (Viner et al., 2012). Adolescents who receive low
levels of support from parents are prone to negative health outcomes, includ-
ing involvement in dating violence (S. Miller, Gorman-Smith, Sullivan,
Orpinas, & Simon, 2009; R. L. Simons, Lin, & Gordon, 1998; Viner et al.,
2012). Future research should examine potential mechanisms linking low
parental support to dating violence. For example, adolescents who perceive
low care from their parents may seek care from another source, including a
dating partner. If adolescents have not had a strong example of love from
their parents, they may be more likely to initially accept and continue to toler-
ate a violent partner.
Associations between low perceived care from teachers or friends and dat-
ing violence victimization were moderate in strength; in comparison, partici-
pation in extracurricular activities was only weakly protective against dating
violence victimization. While enhancing the ability of teachers to support
adolescents, fostering the development of healthy friendships, and increasing
Earnest and Brady 399

adolescents’ participation in extracurricular activities may offer some protec-

tion against experiences of dating violence, the present findings suggest that
these are not the most influential targets for prevention and intervention pro-
grams designed to curb dating violence. Safety in the home and school and
perceived care from parents are likely to be more influential targets.

Stress-Buffering Versus an Additive Model of Risk and

Protection
The present study examined whether perceived care and extracurricular
activity involvement attenuated (weakened) the effects of family violence
and perceived school safety on dating violence victimization. While the
hypothesis was confirmed and attenuating effects were found for most inter-
action tests, attenuating effects were modest compared with observed main
effects, which were comparatively strong in magnitude. Regardless of
resources for support, stressors such as feeling unsafe at school, being a vic-
tim of physical violence by an adult in the home, and witnessing intra-famil-
ial physical violence were strongly associated with likelihood of being
victimized by dating violence. The resources for support examined in this
study did not offer a stress-buffering benefit to the extent that they could
“take away” or greatly reduce the risk conferred by stressor exposure. The
large sample size helped detect relatively modest stress-buffering effects.
One potential reason for smaller than expected stress-buffering effects is
the manner in which social support was measured. Extracurricular activity
involvement may have indicated an adolescent’s level of social connectedness
as opposed to the extent to which specific individuals served as a resource for
support, particularly with respect to navigating dating relationships. Cohen
and Wills (1985) have argued that structural measures of support that assess
the existence of relationships, but do not delve further into the qualities of
those relationships, are less likely to demonstrate stress-buffering effects com-
pared with measures that assess specific qualities of a supportive relationship
that may be advantageous in the presence of a stressor. Extracurricular
involvement is a structural measure of support. Perceived care, which is an
important component of emotional support, may not be particularly advanta-
geous in ameliorating the impact of domestic violence exposure or attending
an unsafe school on dating violence victimization. Informational support (e.g.,
assistance in defining healthy relationships) and appraisal support (e.g., assis-
tance in identifying unhealthy relationships) may be more likely than emo-
tional support to buffer the impact of stressor exposure on dating violence
victimization. It is also possible that parents and other individuals can care for
400 Journal of Interpersonal Violence 31(3)

adolescents while still establishing and reinforcing unhealthy norms with

respect to intimate partner relationships. Indeed, L. G. Simons et al. (2012)
found that adolescents’ perceptions of parental warmth exacerbated—not
attenuated—adolescents’ vulnerability to dating violence victimization in the
presence of family dysfunction. Future research should investigate whether
adolescents who perceive high care from their parents and who have also
experienced violence in the home may perceive abuse in a relationship as part
of caring. Adolescents in this circumstance may learn to make excuses for
their partner and give “second chances,” complicating their views of what
constitutes a healthy relationship or a healthy expression of care.
While perceived care and extracurricular activity involvement only mod-
estly attenuated links between risk factors and dating violence victimization, it
is important to note that these resources for support had direct protective effects
on the likelihood of dating violence victimization. These direct protective
effects were still present when adjusted for exposure to unsafe schools and
family violence in an additive model of dating violence victimization. Feeling
cared for by parents was especially advantageous for adolescents. The present
set of findings is thus more consistent with an additive model of risk and pro-
tection rather than a stress-buffering model in which specific protective factors
can eliminate or greatly reduce the risks associated with stressors.

Strengths and Limitations

The large statewide sample utilized in the present study is a strength, as it
afforded high power to detect small differences between groups. A second
strength is the broad definition of violence in the home, which included wit-
nessed abuse perpetrated on anyone in the home. This definition is not lim-
ited to acts of violence between parent(s) and their partners, and thus allows
for diversity in the composition of families. Much of the literature focuses on
intimate partner violence and may miss other types of violence occurring in
the home.
Limitations of the present study include the cross-sectional survey design,
which prevents inferences about causality, reliance on self-report, and lack of
assessment with respect to dating history of participants or frequency of dif-
ferent types of violence exposure. In addition, the present study used single
items to assess each construct rather than established scales with demon-
strated reliability; however, items have strong face validity.
There were significant differences on most study variables between those
students excluded from the study for having incomplete data and those
included in the study. A greater percentage of those excluded from the study
had reported dating violence victimization and risk factors associated with
Earnest and Brady 401

dating violence victimization in comparison with the study sample; this may
have led to an underestimate of effect sizes. While the current sample is most
likely representative of public high school students in the Midwest, consis-
tency of findings with other research suggests that the associations described
in this study would be generalizable across the United States.

Conclusion
Experience of violence in the home, perceived lack of safety at school, and
low perceived care from parents were strongly associated with dating vio-
lence victimization among adolescents. This is gravely concerning from a
youth development perspective, as 9% of adolescents reported being a victim
of physical violence by an adult in the home, 11% reported witnessing intra-
familial physical violence, 6% reported feeling unsafe at school, and 10%
reported low perceived care from parents. Perceived care and extracurricular
activity involvement appear to only have modest buffering effects on the
relationships between violence in the home or perceived lack of safety at
school and dating violence victimization. The results of this research are most
consistent with an additive model of risk and protective factors in relation-
ship to dating violence victimization rather than a stress-buffering model.
Further research on the stress-buffering hypothesis, including measurement
of different types of support (e.g., informational support, appraisal support),
is recommended.
Family violence and sources of social support are two of many dynamic
factors that may contribute to risk of or protection from adolescent dating
violence. Forces broader than the home and school settings, including cul-
ture, media, and poverty, also influence attitudes toward and experience of
violence (Connolly et al., 2010; Jain et al., 2010). Family and school environ-
ments encompass the settings in which adolescents are likely to spend most
of their time, thus making these environments prime targets for prevention
and intervention.
Health promotion efforts to prevent dating violence victimization among
adolescents should target the individual risk and protective factors examined.
Schools should focus on creating environments where all students feel safe.
Health professionals should continue efforts to end violence in the home, as
this may have the greatest effect in reducing dating violence victimization.
For adolescents who have already experienced violence in the home, efforts
to prevent dating violence should focus on increasing resources for support
and changing norms that support acceptance or tolerance of violence. Parents
should be encouraged and assisted in their efforts to be supportive and
engaged in the lives of their adolescents.
402 Journal of Interpersonal Violence 31(3)

Research efforts should attempt to understand the mechanisms by which a

violent home or an unsafe school environment contribute to risk of dating
violence victimization, the specific aspects of social support that protect
against dating violence victimization, and the qualitative characteristics of
extracurricular activities that are likely to prevent dating violence. Prospective
research studies, which are currently few in number in the dating violence
literature, would aid in this endeavor and inform future prevention efforts.

Acknowledgments
The first author thanks J. Michael Oakes, PhD, for his advice regarding methodologi-
cal decisions, and Diann Ackard, PhD, for her support and introduction to the realm
of adolescent dating violence prevention. The authors acknowledge Ann Kinney,
PhD, Minnesota Center for Health Statistics, Minnesota Department of Health, for her
assistance in procuring the data set and answering questions regarding data collection
and cleaning procedures.

Authors’ Contributions
Alicia A. Earnest conceived of the study, designed the study, performed the data anal-
ysis, and drafted the manuscript. Sonya S. Brady contributed to the study design, data
interpretation, and editing of the manuscript. Both authors read and approved the final
manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publi-
cation of this article.

References
Aldarondo, E., & Sugarman, D. B. (1996). Risk marker analysis of the cessation and
persistence of wife assault. Journal of Consulting and Clinical Psychology, 64,
1010-1019.
Bandura, A. (1971). Social learning theory. New York, NY: General Learning Press.
Barber, B. L., Eccles, J. S., & Stone, M. R. (2001). Whatever happened to the
jock, the brain, and the princess? Young adult pathways linked to adolescent
activity involvement and social identity. Journal of Adolescent Research, 16,
429-455.
Beran, T. N., & Tutty, L. (2002). Children’s reports of bullying and safety at school.
Canadian Journal of School Psychology, 17, 1-14.
Earnest and Brady 403

Breiding, M. J., Black, M. C., & Ryan, G. W. (2008). Prevalence and risk factors
of intimate partner violence in eighteen U.S. states/territories, 2005. American
Journal of Preventive Medicine, 34, 112-118.
Centers for Disease Control and Prevention. (2012). Youth risk behavior surveil-
lance—United States, 2011. Morbidity and Mortality Weekly Report, 61(SS-4),
1-162.
Chiodo, D., Wolfe, D. A., Crooks, C., Hughes, R., & Jaffe, P. (2009). Impact of sexual
harassment victimization by peers on subsequent adolescent victimization and
adjustment: A longitudinal study. Journal of Adolescent Health, 45, 246-252.
Cohen, S., & Wills, T. A. (1985). Stress, social support, and the buffering hypothesis.
Psychological Bulletin, 98, 310-357.
Connolly, J., Friedlander, L., Pepler, D., Craig, W., & Laporte, L. (2010) The ecology
of adolescent dating aggression: Attitudes, relationships, media use, and socio-
demographic risk factors. Journal of Aggression, Maltreatment & Trauma, 19,
469-491.
Cooley, V. E., Henriksen, L. W., Van Nelson, C., & Thompson, J. C. (1995). A study to
determine the effect of extracurricular participation on student alcohol and drug use
in secondary schools. Journal of Alcohol and Drug Education, 40, 71-87.
Cunradi, C. B., Caetano, R., & Schafer, J. (2002). Socioeconomic predictors of inti-
mate partner violence among White, Black, and Hispanic couples in the United
States. Journal of Family Violence, 17, 377-389.
Eccles, J. S., & Barber, B. L. (1999). Student council, volunteering, basketball, or
marching band: What kinds of extracurricular involvement matters? Journal of
Adolescent Research, 14, 10-43.
Eccles, J. S., Barber, B. L., Stone, M., & Hunt, J. (2003). Extracurricular activities and
adolescent development. Journal of Social Issues, 59, 865-889.
Egeland, B. (1993). A history of abuse is a major risk factor for abusing the next gen-
eration. In R. J. Gelles & D. R. Loseke (Eds.), Current controversies on family
violence (pp. 197-208). Newbury Park, CA: Sage.
Espelage, D. L., & Holt, M. K. (2007). Dating violence & sexual harassment across
the bully-victim continuum among middle and high school students. Journal of
Youth and Adolescence, 36, 799-811.
Field, C. A., & Caetano, R. (2005). Intimate partner violence in the US general pop-
ulation: Progress and future directions. Journal of Interpersonal Violence, 20,
463-469.
Foshee, V. A., Benefield, T. S., Ennett, S. T., Bauman, K. E., & Suchindran, C.
(2004). Longitudinal predictors of series physical and sexual dating violence vic-
timization during adolescence. Preventative Medicine, 39, 1007-1016.
Fredricks, J. A., & Eccles, J. S. (2006). Is extracurricular participation associated
with beneficial outcomes? Concurrent and longitudinal relations. Developmental
Psychology, 42, 698-713.
Gagne, M. H., Lavoie, F., & Hebert, M. (2005). Victimization during childhood and
revictimization in dating relationships in adolescent girls. Child Abuse & Neglect,
29, 1155-1172.
404 Journal of Interpersonal Violence 31(3)

Glew, G. M., Fan, M., Katon, W., & Rivara, F. P. (2008). Bullying and school safety.
Journal of Pediatrics, 152, 123-128.
Halpern, C. T., Spriggs, A. L., Martin, S. L., & Kupper, L. L. (2009). Patterns of
intimate partner violence victimization from adolescence to young adulthood in
a nationally representative sample. Journal of Adolescent Health, 45, 508-516.
Jain, S., Buka, S., Subramanian, S. V., & Molnar, B. E. (2010). Neighborhood pre-
dictors of dating violence victimization and perpetration in young adulthood: A
multilevel study. American Journal of Public Health, 100, 1737-1744.
Kalmuss, D. (1984). The intergenerational transmission of marital aggression. Journal
of Marriage and the Family, 46, 11-19.
Laporte, L., Jiang, D., Pepler, D. J., & Chamberland, C. (2011). The relationship
between adolescents’ experience of family violence and dating violence. Youth
& Society, 43, 3-27.
Larson, R. W., Hansen, D. M., & Moneta, G. (2006). Differing profiles of devel-
opmental experiences across types of organized youth activities. Developmental
Psychology, 42, 849-863.
Levendosky, A. A., Huth-Bocks, A., & Semel, M. A. (2002). Adolescent peer rela-
tionships and mental health functioning in families with domestic violence.
Journal of Clinical Child & Adolescent Psychology, 31, 206-218.
Maas, C. D., Fleming, C. B., Herrenkohl, T. I., & Catalano, R. F. (2010). Childhood
predictors of teen dating violence victimization. Violence and Victims, 25, 131-149.
Mahoney, J. L., & Cairns, R. B. (1997). Do extracurricular activities protect against
early school dropout? Developmental Psychology, 33, 241-253.
Mahoney, J. L., & Stattin, H. (2000). Leisure activities and adolescent antisocial
behavior: The role of structure and social context. Journal of Adolescence, 23,
113-127.
Malik, S., Sorenson, S. B., & Aneshensel, C. S. (1997). Community and dating vio-
lence among adolescents: Perpetration and victimization. Journal of Adolescent
Health, 21, 291-302.
Marsh, H. W., & Kleitman, S. (2003). School athletic participation: Mostly gain with
little pain. Journal of Sport & Exercise Psychology, 25, 205-228.
Masten, A. S., Cutuli, J. J., Herbers, J. E., & Reed, M. J. (2009). Resilience in devel-
opment. In S. J. Lopez & C. R. Snyder (Eds.), Oxford handbook of positive psy-
chology (2nd ed., pp. 117-131). New York, NY: Oxford University Press.
Masten, A. S., & Powell, J. L. (2003). A resilience framework for research, policy,
and practice. In S. S. Luthar (Ed.), Resilience and vulnerability: Adaptation in
the context of childhood adversities (pp. 1-26). Cambridge, MA: Cambridge
University Press.
McNeal, R. B. (1995). Extracurricular activities and high school dropouts. Sociology
of Education, 68, 62-81.
Miller, E., Breslau, J., Chung, W. J. J., Greif Green, J., McLaughlin, K. A., & Kessler,
R. C. (2011). Adverse childhood experiences and risk of physical violence in
adolescent dating relationships. Journal of Epidemiology and Community Health,
65, 1006-1013.
Earnest and Brady 405

Miller, K. E., Sabo, D., Farrell, M. P., Barnes, G. M., & Melnick, M. J. (1999). Sports,
sexual behavior, contraceptive use, and pregnancy among female and male high
school students: Testing cultural resource theory. Sociology of Sport Journal, 16,
366-387.
Miller, S., Gorman-Smith, D., Sullivan, T., Orpinas, P., & Simon, T. R. (2009).
Parent and peer predictors of physical dating violence perpetration in early ado-
lescence: Tests of moderation and gender differences. Journal of Clinical Child
& Adolescent Psychology, 38, 538-550.
O’Keefe, M., & Treister, L. (1998). Victims of dating violence among high school
students. Are the predictors different for males and females? Violence Against
Women, 4, 195-223.
O’Leary, K. D., Smith Slep, A. M., Avery-Leaf, S. A., & Cascardi, M. (2008). Gender
differences in dating aggression among multiethnic high school students. Journal
of Adolescent Health, 42, 473-479.
Pollard, J. A., Hawkins, J. D., & Arthur, M. W. (1999). Risk and protection: Are
both necessary to understand diverse behavioral outcomes in adolescence? Social
Work Research, 23, 145-158.
Simons, L. G., Simons, R. L., Lei, M., Hancock, D. L., & Fincham, F. D. (2012).
Parental warmth amplifies the negative effect of parental hostility on dating vio-
lence. Journal of Interpersonal Violence, 27, 2603-2626.
Simons, R. L., Lin, K., & Gordon, L. C. (1998). Socialization in the family of origin
and male dating violence: A prospective study. Journal of Marriage and Family,
60, 467-478.
Spriggs, A. L., Halpern, C. T., Herring, A. H., & Schoenbach, V. J. (2009).
Family and school socioeconomic disadvantage: Interactive influences on
adolescent dating violence victimization. Social Science & Medicine, 68,
1956-1965.
Tajima, E. A., Herrenkohl, T. I., Moylan, C. A., & Derr, A. S. (2011). Moderating the
effects of childhood exposure to intimate partner violence: The roles of parenting
characteristics and adolescent peer support. Journal of Research on Adolescence,
21, 376-394.
Teten, A. L., Ball, B., Valle, L. A., Noonan, R., & Rosenbluth, B. (2009).
Considerations for the definition, measurement, consequences, and prevention
of dating violence victimization among adolescent girls. Journal of Women’s
Health, 18, 923-927.
Tjaden, P., & Thoennes, N. (2000). Extent, nature, and consequences of intimate
partner violence: Findings from the National Violence Against Women Survey.
Washington, DC: U.S. Department of Justice.
Tyler, K. A., Brownridge, D. A., & Melander, L. A. (2011). The effect of poor parent-
ing on male and female dating violence perpetration and victimization. Violence
and Victims, 26, 218-230.
Viner, R. M., Ozer, E. M., Denny, S., Marmot, M., Resnick, M., Fatusi, A., & Currie,
C. (2012). Adolescence and the social determinants of health. The Lancet, 379,
1641-1652.
406 Journal of Interpersonal Violence 31(3)

Windle, M., & Mrug, S. (2009). Cross-gender violence perpetration and victimization
among early adolescents and associations with attitudes toward dating conflict.
Journal of Youth & Adolescence, 38, 429-439.
Youniss, J., Yates, M., & Su, Y. (1997). Social integration: Community service
and marijuana use in high school seniors. Journal of Adolescent Research, 12,
245-262.

Author Biographies
Alicia A. Earnest, MPH, received her MPH in Maternal and Child Health from the
University of Minnesota School of Public Health. Her primary academic interests
include dating violence prevention, health disparities, social determinants of health,
and the influence of culture on health.
Sonya S. Brady, PhD, is an Associate Professor within the University of Minnesota
School of Public Health, Division of Epidemiology and Community Health. Her
research examines the influence of psychosocial factors on health behaviors among
youth, including mechanisms that link stressful life circumstances to risk behavior
and factors promoting resilience.